There seems to be endless goalpost moving around this though. Do you release as soon as you possibly can and learn and validate, or release a relatively polished product and have a strong initial base of customers help finesse it?

It's also kind of hard to get meaningful feedback without some kind of big PR push.

The way I approach it is as a pyramid. First version, really shitty, I just need to get few customers/users and get him/her interested. if they see big holes, go fix it and come back. Work on fixing smaller holes and other features, collect feedback from a slightly larger audience. Repeat such feedback/iteration cycles.

As product becomes more mature, increase size until you have a 1.0, at which point you have a loyal fan base that loves the product and is willing to pay. Then you go full PR, market like crazy and burn VC money in growth.

The point is at each stage you are still validating and gathering feedback without the risk of losing out on a large number of customers.

>It's also kind of hard to get meaningful feedback without some kind of big PR push.

In my experience, you can get meaningful feedback by watching one, or ten, or a hundred users use your app. Indeed, given the level of tailoring that you can (and should) do to pick the first batch of users, you should be able to give them very close attention

If you don't have evangelists in your early-alpha userbase, then something is probably wrong.